The superstitious atheist

Justin Taylor today pointed out an article in the Wall Street Journal entitled Look Who's Irrational Now:

The reality is that the New Atheist campaign, by discouraging religion, won't create a new group of intelligent, skeptical, enlightened beings. Far from it: It might actually encourage new levels of mass superstition. And that's not a conclusion to take on faith -- it's what the empirical data tell us.

"What Americans Really Believe," a comprehensive new study released by Baylor University yesterday, shows that traditional Christian religion greatly decreases belief in everything from the efficacy of palm readers to the usefulness of astrology. It also shows that the irreligious and the members of more liberal Protestant denominations, far from being resistant to superstition, tend to be much more likely to believe in the paranormal and in pseudoscience than evangelical Christians.

The article also notes the higher-educated people are more likely to believe paranormal phenomena.

My cynical side notes that this study comes from Baylor, a baptist-affiliated university. On the other hand, the actual polling was conducted by Gallup and the surveys they used are available online.

Update: The scores for paranormal vs. belief and church attendance can be found on pages 51-52 (49-50 if you follow the numbers at the bottom of the page) of this PDF.

"Catholics", "black protestants" (does evangelical imply white?), and "mainline protestants" all scored higher for belief in paranormal phenomena than those who affiliated themselves with no religious group. Comparing to those in the "Other" category seems a little silly, as this group is basically by definition made up of those who don't fit into typical categories.

Figure 20 measures church attendance versus belief in the paranormal. It seems as though the authors of the paper must have read a paper I saw on "How to lie with statistics" as the y-axis of the graph only shows the interval 22 - 30, magnifying the apparent differences. The actual difference is only 11% and 16% when comparing weekly church attenders with, respectively, the average score and the score of those who never attended church.